A customer involved in research on image processing using machine learning asked us about a machine that runs the computational library "JAXopt" that performs GPU processing at high speed.
We also plan to use 3DCG software and Adobe products, and would like a configuration suitable for these purposes within a budget of 100 million yen.
When selecting specs, we have heard that you would like to prioritize GPU and memory capacity, taking into account the nature of the application, and specifically, we are assuming an NVIDIA GPU with a VRAM capacity of 20GB or more.
In addition, on the Windows machine that is currently being used, JAXopt is running by installing jaxlib in combination with the Anaconda environment, and if possible, it is recommended to pre-install JAXopt and CUDA Toolkit on the new machine. Thank you for your request.
Based on these requests, we have considered the following configuration.
CPU | Intel Core i9-14900K (3.20GHz 8 cores + 2.40GHz 16 cores) |
memory | 128GB |
storage | 1TB SSD M.2 NVMe Gen4 |
video | NVIDIA Geforce RTX4090 24GB |
network | on board(2.5GBase-T x1) Wi-Fi x1 |
Housing + power supply | Middle tower type housing + 1500W |
OS | Microsoft Windows 11 Professional 64bit |
Others | CUDA Toolkit/JAXopt installation |
A configuration that emphasizes VRAM capacity and memory capacity within the budget
This configuration is equipped with the latest 2024th generation Intel Core series high-end model Intel Core i4-14K as of April 9.
For the GPU, we selected the NVIDIA Geforce RTX4090 24GB, which has the highest VRAM capacity among the products available within our budget.
In addition, in terms of memory capacity, it is equipped with 128GB, which is the maximum capacity that can be installed on the motherboard with this configuration.
With 24GB of VRAM capacity and 128GB of memory capacity, you can expect high processing speed when using jaxopt's GPU processing.
jaxopt・Pre-installation of CUDA Toolkit is also possible
It is also possible to ship with JAXopt and CUDA Toolkit pre-installed.
When installing the CUDA Toolkit on Windows, it is required to install Microsoft Visual Studio at the same time, so it is assumed that Visual Studio Community, which is available for free, will be installed.
Also, since you said that you already have an Anaconda virtual environment in use, in order to install JAXopt, please send us the exported data of the virtual environment, import it to a new machine, and duplicate it. We have proposed a way to do this.
The configuration of this case study is based on the conditions given by the customer.
We will flexibly propose machines according to your conditions, so please feel free to contact us even if you are considering different conditions than what is listed.
■ Keywords・What is JAXopt?
|
■ Click here for details and inquiries about this PC case * Please enter the name of the case or your desired conditions. |